Email Spam Classifier – SVM

A spam/ham email classifier built with Support Vector Machines (SVM) and a Bag of Words text representation. Also submitted to Kaggle — see eladmoshe98.

Approach

• Text preprocessing: tokenization, stop-word removal, stemming
• Feature extraction: Bag of Words (CountVectorizer / TF-IDF)
• Primary classifier: SVM (linear kernel)
• Comparison classifiers: Naive Bayes, Logistic Regression, Random Forest

Running

pip install scikit-learn pandas numpy jupyter
jupyter notebook spam_classifier.ipynb

Tech Stack

Python · scikit-learn · pandas · NLTK